I'm not sure I fully understand the question - I know who the big players are in the microarray field (probably most microarray studies deal with cancer) but wouldnt neceessarily what are the best informatics labs. Feel free to PM me...

---

_______________________________________________

I know several people at Boston University looking at the molecular signals leading to environmental-induced breast cancer (BU also has an entire Bioinformatics department). If their research is along the lines of what you'd be looking at, you can look up their research and cross-reference for microarrays. If that is not what you are looking for, I can try to get you a list of names, but they are hard to track down in the summer (damn professors!). I cannot speak to the "who's-who" factor, but it may be a place to start.

---

====================================

Life isn't measured by the number of breaths we take, but by the number of moments that take your breath away.

pretty much any studies involving DNA microarrays is of interest (besides the rather trivial approach of using neural network on DNA ma.)
I'll look at BU. thanks

PM me if you need specific names. Be sure to check out the School of Public Health/Environmental Health/Sherr Lab.

---

====================================

Life isn't measured by the number of breaths we take, but by the number of moments that take your breath away.

You might be able to get a list of projects/contacts from the TARP folks at the NCI.

http://ccr.cancer.gov/tech_initiatives/tarp/

crap...don't know why I didn't look there first...I searched NIH a bit and forgot to check NCI.
thanks for the help.

[reply]Working on DNA microarrays as models to represent gene expression, comparative gene expression etc.
I am particularly interested in being able to represent gene interactions (first degree, which is basically the 'importance' of a gene for a specific biological process, second degree that is positive and negative synergies between genes, and possibly third degree although the complexity of dealing with DNA microarrays in this case becomes bad)...
Potential applications are identifying the key genes for specific cancers and their interaction, better cancer classification, and hopefully better diagnosis/treatment/prognosis.

I can find the relevant work done in this area from a bioinformatics perspective (close to nil beyond first degree) but I'm not familiar enough with the who's who in oncology/biology etc to find out the truly relevant work in this area from a strictly biological perspective (to be able to compare our models with what is really observed)...[/reply]

Hi Francois,
you know there's a reason that you find so little published in that area. Trying to infer biological meaning (beyond first degree) from microarray data is incredibly difficult on many levels. Starting from the fact that microarray results always only paint an incomplete picture (you'll always have transcripts that will regularly escape detection, no matter which system you use), to the fact that a lot of biologically important interactions between gene products do not manifest themselves in altered transcript levels, to the fact that large scale wet lab experimentation to verify predictions from microarray analyses would be so time- and cost-consuming that it's not really realistic. I would say your best bet would be to calibrate your models on some of the well-known metabolic or signaling pathways, where knowledge from many years of experimentation and from many different labs has been compiled (see for example the KEGG database at http://www.genome.ad.jp/kegg/metabolism.html). If your model can correctly predict/reproduce some of those interactions, your probably on the right track.
Keep me posted about your progress, I'd be really interested too!

Cheers,
malte

This is precisely what I am after. I don't mind whether the results come from microarrays or more traditional biological experiments. As a matter of fact, I'd prefer to have some bio. exp. stating that gene A and B combined together have a synergistic effect to affect gene C (not sure if these types of results are difficult to obtain in strict biological setting though) and use this as benchmarks to compare with what our models look like.
I wouldn't want to compare models inferring knowledge from DNA microarrays (what I am doing) with other models doing the same thing. Results would be rather meaningless, specially considering the issues with DNA microarray quality control.

I just looked at the KEGG DB and this is exactly what I was after. This is going to be very useful. Thanks a lot.

Man I feel stupid! 10 years ago I was taking classes like Eukaryotic Molecular Genetics (did I even spell that right?) and Biochemistry of Nucleic Acids and various molecular biology laboratory classes (DNA sequencing, PCR, plasmid vectors etc..). Now I remember about 1% of that stuff. I have been playing Naval Officer for 10 years now (absolutely NO regrets!). It is amazing how much you forget if you don't use it every day. I was looking at my general exams recently and I don't really even understand what I wrote! I was pretty smart back then! Sometimes I feel like I should just turn in my PhD. Of course the fact that I get an hour or more to work out on the clock every day, 30 days of paid leave and am making quite a bit more money than any of my former professors makes me feel a bit better. My work day today will consist of hopping in a helicopter and flying to Alabama to get some lunch, then we will probably fly up and down the coast looking for sharks. Not a bad way to spend the day! That sounds like interesting stuff Francois! Wish I knew what you were talking about! Good luck.

Mike P.

This is very interesting stuff to me anyway...and knowing that it may actually help people make it a lot more interesting...
Where I feel stupid is when I realize that I went to Academia because everyone said 'oh, your time there will be so much easier than the industry, great working hours, not too much work etc.' I guess none of these guys actually worked in a research school that also wants you to teach...publish, write grants, get funding, teach (2 classes a semester, half the time without a TA), advise students, mentor master and phd students etc...like 5 jobs in one...I wish I could focus on research which I guess is only likely in the industry where the hours are so crazy...man I'll take the 40-45hrs anytime over the 55-60hrs with work flexibility...can't remember the last time I went somewhere on holidays without work to do.

So this is how you're trying to cure your Achilles problem huh?

You're going to clone yourself! A New & Improved Francois....yikes.

: P

---

_______________________________________________________

Seeing him in deep torment, I said. " You can have my last GU , but its been down my pants for most of the run". - John Hirsch

Take care of your body, its by far the coolest thing you're going to ever own.- Can't remember who told me this, but I love it.

yeah! Twice as big! so I'll reach 5 feet! ;-)

Francois

I hear you. In graduate school I spent countless evenings and weekends in the lab and the library. I ALWAYS had some reading or writing that I needed to do. I could never really relax during my time off. My first year in the Navy, I just couldn't wind down from that pace. No problem now. I can't tell you how nice it is to leave work on Friday and really have nothing on my mind and nothing hanging over my head all weekend. I have plenty of time off (30 days leave plus all of those federal holidays). I am glad I went Navy rather than academia. Sometimes I miss working in my field though. When I retire from the Navy, I will find some sleepy little junior college and teach part time. I wouldn't take a teaching position at a large research university even if I got good offers. Not for me any more.

Mike

Do you have to be a US national to join the Navy? ;-)

[quote]This is precisely what I am after. I don't mind whether the results come from microarrays or more traditional biological experiments. As a matter of fact, I'd prefer to have some bio. exp. stating that gene A and B combined together have a synergistic effect to affect gene C (not sure if these types of results are difficult to obtain in strict biological setting though) [/quote]

You bet! That's why I'm saying, you'll likely never see large scale wet lab experimentation just to verify microarray results.

[quote]I wouldn't want to compare models inferring knowledge from DNA microarrays (what I am doing) with other models doing the same thing. Results would be rather meaningless, specially considering the issues with DNA microarray quality control.[/quote]

You're speaking right from my heart :-)
(And this is coming from a biologist with 5+ years experience working with microarrays)

[quote]I just looked at the KEGG DB and this is exactly what I was after. This is going to be very useful. Thanks a lot.[/quote]

Glad I could help.
Like I said, let me now if you develop some neat new models, I'd be very interested.

Cheers,
malte

Francois

No you dont. Not sure what the process is but we have a few Brits and others.

Mike

You need Gene.

Are you Gene?

No, this is Gene.

Gene is a computer?

The fastest there is...

Ohmigawd, the Tour corrupted my brain with useless advertising!

too much titanium in your brain I believe... ;-)